The goal of the work proposed is to identify the fundamental sensorimotor transformations and predictive mechanisms underlying the human capacity to coordinate simultaneous whole body and limb movements, which is compatible with the core objective of the NIH/NINDS Sensorimotor Integration Study Section. In typical everyday activities, healthy people reach fluidly and accurately despite large variations in the dynamics correlated with their body motion, and they simultaneously maintain balance despite large reciprocal forces on the body axis from the moving arm. The challenge which arm-body coordination poses to the CNS is evident in the many clinical conditions that produce specific deficits in control of multi-joint turning behaviors, including labyrinthine and cerebellar deficit, Parkinson's disease, and diabetic neuropathy. Multi-joint coordination matures gradually in typically developing children and declines variably in the elderly. Falls during turning often lead to serious injury. Our project focuses on specifying the signals used in normal compensation for self-generated Coriolis forces generated between the torso turn and reach for an object and the roles of the cerebellum and labyrinth in these In order to help develop strategies for prevention and treatment of these problems. We will 1) experimentally test three models describing the interaction between sensorimotor transformations subserving dynamic motor control of the torso rotation in space and of the arm relative to the torso, 2) decompose the predictive from the reactive components of muscle activity while subjects learn to compensation for novel Coriolis forces, 3) resolve the relative contributions of vestibular and proprioceptive signals to sensorimotor calibration of torso rotation in space and arm-torso control and 4) define the role the cerebellum normally plays in compensation for self-generated Coriolis forces and alternative modes of compensation when cerebellar function is compromised. Our approach employs two unique laboratory facilities for manipulating Coriolis forces, a rotating room and a servo platform for altering the relationship between intended and inertial torso rotation. The approach spans biomechanical, physiological, clinical and computational domains. A physiologist-engineer and a neurorehabilitation physician are collaborating on the project with the two PIs. Since its inception in 2003, this project has included postdoctoral fellows with degrees in biomechanics, neuroscience and engineering, students who have earned PhDs in Neuroscience, Physics and Psychology, and currently have PhD, and undergraduate students from all three of these domains working in the laboratory. The proposed budget includes support for one PhD student.